Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.</em>—The Washington Department of Fish and Wildlife and Tribal co-managers are using the Ecosystem Diagnosis and Treatment (EDT) model to identify the spatial and temporal habitat limits of salmon populations and predict the effects of proposed habitat restoration projects for ESA-listed Chinook salmon <em>Oncorhynchus tshawytscha </em>in two Puget Sound watersheds. The collaborative, iterative process focused on habitat-based population models for the Dungeness and Dosewallips watersheds. Workshops were held to develop quantitative characteristics of current, historic, hypothetical properly functioning, and future habitat conditions. The model predicted salmon populations in the watersheds for each set of habitat conditions. Recovery targets were based on the predicted populations for historic and hypothetical properly functioning conditions. Future populations were modeled using projected habitat conditions with individual habitat restoration and protection actions already proposed and combinations of these actions. Populations resulting from further habitat degradation were estimated using the effects of projected human population growth on habitat.

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.</em>—Natural resource management requires difficult decisions, broad societal costs, and sacrifices from private landowners and public agencies. With so many financial, ecological and cultural resources at stake, policy-makers, managers, and citizens need scientific predictions that can help resolve conflicts and balance the often competing needs of ecosystems and communities. Modeled information is essential for meeting this need. The words “model uncertainty” are often misinterpreted as describing a lack of knowledge about model output. In fact, they describe knowledge, not only of the one most likely modeled estimate, but also of all the other possible estimates that the model might have provided, and their likelihood. We present six case studies, from salmon habitat recovery planning, illustrating how scientists can provide more useful products by describing distributions of possible outcomes as formal probability distributions, as confidence intervals, or as descriptions of alternative scenarios. In terms of management effectiveness, the communication and use of model uncertainty can be at least as important as the quality of the original model.

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.</em>—Past attempts to improve population models of Pacific salmon <em>Oncorhynchus </em>spp. by adding indices of freshwater or marine conditions have shown mixed success. To increase chances that such models will remain reliable over the long term, we suggest adding only environmental covariates that have a spatial scale of positive correlation among monitoring locations similar to, or greater than, that of the salmon variables that scientists are trying to explain. To illustrate this approach, we analyzed spawner and recruit data for 120 populations (stocks) of pink <em>O. gorbuscha</em>, chum <em>O. keta</em>, and sockeye <em>O. nerka </em>salmon from Washington, British Columbia, and Alaska. Salmon productivity of a given species was positively correlated across stocks at a spatial scale of about 500–800 km. Compared to upwelling and sea-surface salinity, summer sea-surface temperature (SST) showed a more appropriate spatial scale of positive covariation for explaining variation in salmon productivity, and was a significant explanatory variable when added to both single-stock and multi-stock spawner-recruit models. This result suggests that future models of these salmon populations should possibly include stock-specific, summer SST. To further explore our understanding of salmon population dynamics, we developed 24 alternative stock–recruitment models. We compared these models in three ways: (1) their fit to all past data, (2) their ability to forecast recruitment, and (3) their performance inside an “operating model,” which included components for dynamics of the natural ecological system, stock assessments based on simulated sampling of data, regulation-setting based on those assessments, and variation in implementing those regulations (reflecting noncompliance or other sources of outcome uncertainty). We also compared single-stock models with multi-stock models (meta-analyses). The latter led to more precise estimates of the effects of SST on log_e(recruits/spawner) and greater accuracy of preseason forecasts for some stocks. Analyses with the operating model show that reducing outcome uncertainty should be a top management priority.

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.—</em>We understand our environment through our senses and tend to interpret the behavior of other animals in the context of the world we understand. Butterflies and flowers sometimes show distinctive patterns in ultraviolet light that are important to them but invisible to us. Likewise, the senses of fish and their experience of the world are very different from ours. Many aspects of a salmon’s environment, such as olfactory stimuli, are completely invisible to us. Other factors, like certain aspects of habitat alteration, are visible but unnoticed because they occurred gradually or long ago. Like Poe’s purloined letter they are cryptic—there for us to see if we only knew what to look for. As we build salmon models we base them on what we understand is important to the fish. However, our anthropocentric bias may cause us to overlook or misinterpret factors of importance. In addition, our necessarily simplified models, when applied to management, may result in a pernicious simplification of the salmon populations we wish to preserve. For example, if we model and manage for a dominant (or highly visible or easily monitored) salmon life history we may inadvertently eliminate other life histories of equal importance, or reduce diversity in ways that affect population viability. We should actively seek to identify important factors missing from our models and be aware of critical assumptions. Recognizing that our models are tools used to understand and manage salmon, we should try to understand the broader implications of these models to the future of the salmon we hope to preserve. In this essay, I offer speculation about what we may be missing in freshwater habitat, life history diversity, metapopulation dynamics, ocean survival, and water chemistry. I also consider the question of scale, and the effect our philosophical viewpoint may have on the direction and application of our modeling efforts and the likelihood of successful outcomes.

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.—</em>Salmon have complex life histories that have been extensively studied, particularly in freshwater, yet most salmon management relies on models that ignore much of salmon life history. For instance, calculation of optimal escapement for most Pacific salmon stocks summarizes their entire life history into a single relationship between spawners and subsequent recruits. Similarly, most analyses of salmon habitat have used models that fail to integrate the complex life history of salmon and have often considered only a single “limiting factor.” Computational methods and models are now being used to incorporate life history and habitat information directly into evaluations of both harvesting and habitat management policies. Challenges and opportunities in using life history models include (1) the need for better dynamic understanding of how habitat affects survival, (2) turning current “expert system” analysis into statistical estimation, (3) application of life history models to hatchery/wild interaction, (4) quantifying essential fish habitat using life history models, (5) using real data and modeling stock structure in evaluation of harvest strategies, and (6) use of such models to explore salmon/ocean interactions.

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.</em>—The Ecosystem Diagnosis and Treatment (EDT) model is being used to build working hypotheses to direct habitat restoration and protection activities in most Pacific Northwest salmon watersheds. The EDT model is used to provide a basis for moving forward with restoration and protection activities, evaluating progress, and refining restoration strategies. The model consists of four components: 1) characterization of the aquatic environment, 2) species-habitat rating rules, 3) life history trajectories, and 4) population performance computations. The environmental characterization is a reach-scale, monthly time step, species-neutral depiction of the stream that focuses on environmental features relevant to salmonids. The species-habitat rating rules are explicit assumptions about the relationship between the stream reach characterization and species-life stage survival. Life history trajectories are multiple computer-generated pathways through the environment. Finally, life history and population performance, defined by Beverton–Holt productivity and capacity parameters, is calculated for each life history trajectory and these trajectories are combined across spatial and biological scales to compute population performance. The model is a freely accessible, web-based tool (http://edt.jonesandstokes.com).

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.—</em>Here we review developments in paleoecological reconstruction of Pacific salmon abundance and discuss the new management context and implications provided by the reconstructions. Currently, two approaches are yielding long term reconstructions of salmon abundance over the last hundreds to thousands of years. First, in sockeye salmon <em>Oncorhynchus nerka </em>nursery lakes, the abundance of adult salmon is reflected in chemical and biological characteristics of lake sediments. These indicators have been used to reconstruct patterns of salmon abundance over 2,500 years and are compared at several points by archeological data. Second, emerging techniques using riparian tree-ring-growth have produced sub-decadal resolution reconstructions of stream-spawning sockeye, Chinook <em>O. tshawytscha</em>, pink <em>O. gorbuscha</em>, and chum <em>O. keta </em>salmon populations over the last 150–350 years. Paleoecological reconstructions provide important insights into salmon abundance and their variability prior to European settlement of western North America. For example, sediment-based reconstructions show periods of naturally low sockeye salmon abundance at ~A.D. 1800 and from ~A.D. 0–700 in Alaskan lakes, and tree-ring based reconstructions show river-specific patterns in abundance with cycles of 21–68 years in duration. Both types of reconstruction also suggest relatively rapid, natural “recovery” of salmon populations after periods of low abundance. As additional reconstructions become available and a more synthetic understanding of them is developed, paleoecological reconstructions will allow better evaluation of management paradigms (e.g., the long-term fidelity of Pacific Decadal Oscillation cycles and regional salmon abundance) as well as identification of additional patterns that cannot be extracted from limited historical data sets. Paleoecological perspectives play a potentially important role in changing societal expectations of salmon resources by recognizing natural variations in abundance. Such expectations, if tempered by acknowledging natural changes in salmon productivity, can be incorporated into flexible models, management and restoration strategies.

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.</em>—The 1996 Sustainable Fisheries Act states that all federal fisheries management plans should contain a description of essential fish habitat (EFH). While much emphasis has been placed on estimating EFH for marine stocks, very little attention has been paid to doing so for Pacific salmon <em>Oncorhynchus </em>spp., in part due to their complex life histories. An earlier assessment of EFH for Pacific salmon across the west coast of the United States focused on the freshwater component of EFH due to limited knowledge about marine distributions. That analysis concluded that a more in-depth and smaller-scale examination was needed to assess how freshwater habitat affects the various life stages. Here we use a detailed life history model for Pacific salmon to estimate the freshwater component of EFH for two threatened populations of Chinook salmon within a large watershed draining into Puget Sound, Washington, USA. By accounting for proposed harvest rates, hatchery practices, and habitat structure, we identified 23 of 50 subbasins as EFH for ensuring no significant decrease in the total number of spawners relative to current average escapement. Our analytical framework could be easily applied to other populations or species of salmon to aid in developing recovery and management plans.

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.</em>—Stream carrying capacity for anadromous salmonids that rear to the smolting stage in freshwater can be predicted from a sequence of cause-response functions that describe fish preferences for macro-habitat features. The channel unit (e.g., pool, glide, riffle) is a useful stratum for quantifying rearing capacity for salmonids, and is a hydrologically meaningful unit for predicting the response of stream morphology to watershed processes. Thus, channel units are the natural link between habitat-forming processes and habitat requirements of salmonids. Maximum densities of juvenile salmonids that can be supported in a channel unit are related to availability of preferred habitat features including velocity, depth, cover, and substrate. Within channel unit types, maximum densities of salmonid parr will shift predictably as availability of cover from wood and boulders increases. Within stream reaches, additional variation in maximum rearing densities can be accounted for by light penetration and nutrient load. As salmonids grow, their habitat preferences change and the preferred habitat associated with their increasing size becomes less and less available. Further, territory size of salmonids increases exponentially with fish length, such that the demand for territory to support surviving members of a cohort increases at least through their first year of life. Changing habitat preferences and space demands, juxtaposed against shrinking habitat availability with the onset of summer low flows often results in a bottleneck to rearing capacity for age >1 salmonids in wadable streams. Habitat measurements in Oregon streams indicate that depths preferred by steelhead (anadromous rainbow trout) <em>Oncorhynchus mykiss </em>become scarce as parr exceed 15 cm in length, which coincides with the approximate threshold length for steelhead smolts. We present a generalized framework, called the Unit Characteristic Method, for accumulating effects of these habitat factors at the channel unit and reach-level scales to estimate carrying capacity for rearing salmonids in a basin. Our subsequent chapter in this book presents a demonstration of how this method can be applied to predicting salmonid production in streams.

Pacific Salmon Environmental and Life History Models: Advancing Science for Sustainable Salmon in the Future

<em>Abstract.</em>—In this review, we argue that modeling salmonid populations needs to integrate multispecies dynamics to better address the keystone role of salmon in their ecosystems and management needs for co-occurring species. Despite several challenges of modeling salmon communities, including multiple threats, variable spatial and temporal scales, and different types of interspecific interactions, a number of modeling approaches are potentially available to examine interspecific interactions. We examine these choices in the context of the salmon communities, and discuss opportunities for improving multispecies models to successfully address the ecological and management questions facing salmon populations. To effectively integrate multiple species into salmonid population models, researchers should utilize relatively simple, transparent models that have the capability of examining parameter uncertainty and systematically adding complexity.